J Refract Surg. 2021 Jan 1;37(1):60-68. doi: 10.3928/1081597X-20201030-02.
To assess the accuracy of intraoperative aberrometry, the Barrett True-K No History (Barrett TKNH), Barrett TKNH with posterior corneal measurements (Barrett TKNH with PC), Shammas-PL, and Haigis-L formulas in patients with cataract who had prior myopic refractive surgery.
This was a retrospective consecutive case series of patients with prior myopic refractive surgery undergoing cataract extraction. Mean absolute error (MAE) and median absolute error (MedAE) of refraction prediction were compared for each formula. Interactions of each biometry measurement were modeled for each formula to evaluate those with the most significant impact on refraction prediction.
One hundred sixteen eyes of 79 patients were analyzed. MAE was 0.40 ± 0.33 diopters (D) for intraoperative aberrometry and 0.42 ± 0.31 D for the Barrett TKNH, 0.38 ± 0.30 D for the Barrett TKNH with PC, 0.47 ± 0.38 D for the Shammas-PL, and 0.56 ± 0.39 D for the Haigis-L formulas. Comparisons between formulas were significant for Barrett TKNH versus Barrett TKNH with PC formulas (P = .046), Barrett TKNH with PC versus Shammas-PL formulas (P = .023), and for all comparisons with the Haigis-L formula (P < .001), and not significant for all other comparisons (P > .05). Eyes were within ±0.50 D of prediction 73%, 72%, 69%, 62%, and 52% of the time for intraoperative aberrometry, the Barrett TKNH with PC, Barrett TKNH, Shammas-PL, and Haigis-L formulas, respectively. Corneal asphericity (Q value) was significantly associated with prediction error for all five methods. Changes in anterior chamber depth had a significant impact on Shammas-PL prediction errors.
Newer technology using information from the posterior cornea modestly improved outcomes when compared to established methods for intraocular lens selection in eyes that had previous laser refractive surgery for myopia. [J Refract Surg. 2021;37(1):60-68.].
评估 Barrett True-K 无历史（Barrett TKNH）、Barrett TKNH 加后角膜测量值（Barrett TKNH with PC）、Shammas-PL 和 Haigis-L 公式在既往近视屈光手术患者中的术中像差测量的准确性。
这是一项回顾性连续病例系列研究，纳入了既往接受过近视屈光手术的行白内障手术患者。比较了每种公式的屈光度预测平均绝对误差（MAE）和中位数绝对误差（MedAE）。对每种公式的每个生物测量指标进行了相互作用建模，以评估对屈光预测影响最大的指标。
纳入了 79 例患者的 116 只眼。术中像差的 MAE 为 0.40±0.33 屈光度（D），Barrett TKNH 为 0.42±0.31 D，Barrett TKNH with PC 为 0.38±0.30 D，Shammas-PL 为 0.47±0.38 D，Haigis-L 为 0.56±0.39 D。Barrett TKNH 与 Barrett TKNH with PC 公式（P=0.046）、Barrett TKNH with PC 与 Shammas-PL 公式（P=0.023）以及与 Haigis-L 公式的所有比较（P＜0.001）差异均有统计学意义，而与其他公式的比较差异均无统计学意义（P＞0.05）。术中像差、Barrett TKNH with PC、Barrett TKNH、Shammas-PL 和 Haigis-L 公式的预测值在±0.50 D 以内的比例分别为 73%、72%、69%、62%和 52%。五种方法的预测误差均与角膜非球面性（Q 值）显著相关。前房深度的变化对 Shammas-PL 预测误差有显著影响。
与既往用于选择人工晶状体的成熟方法相比，使用来自后角膜信息的新技术，在既往接受过近视激光屈光手术的患者中，对选择人工晶状体的结果有适度改善。[J Refract Surg. 2021;37(1):60-68.].
